Methodology & Disclosure

Atlas helps identify market regime, directional bias, confidence, tradeability, and execution mode. It does not guarantee profit and it is not a standalone buy/sell signal.

Always check Data Status and Daily Score Date first. If the dashboard is stale, use it only as historical context until the next successful daily update.

Mode overrides the raw score. If Mode says NO_TRADE, stand aside even if the Atlas Score looks directional. This is the defensive layer that should protect against poor-quality macro-shock setups.

When institutional data exists for a market, Atlas combines weekly positioning with price structure and geometry.

If CFTC is missing, Atlas falls back to price structure and geometry only. This is useful for display and testing, but it should be treated as weaker than CFTC-confirmed scoring.

COT/TFF is the official futures-positioning source. It is useful for institutional crowding and swing bias, not intraday timing. The history page records each active COT z-score beside later market movement.

Yahoo Finance is a temporary public feed for V1 testing. Production-grade scoring should eventually come from the same MT5/CMC environment used for execution.

CME-listed futures and options are useful for exchange-traded validation. FX gamma should be labeled listed-options-only unless OTC data is available.

OFR Hedge Fund Monitor is a weekly or slower macro-risk context layer. It should not be used as a live signal, but it can help identify systemic leverage and hedge-fund crowding conditions.

Daily Atlas controls the main score, regime, mode, confidence, and tradeability. It should run after the U.S. trading day closes.

Intraday Atlas uses hourly data to label wash/rinse session timing: Asian, European, or U.S. It does not replace the Daily Atlas score.

Wash and rinse means price may have swept liquidity or created a false break. Directional change still requires reclaim, retest, and structure break.

Confidence measures component agreement, data availability, trend strength, regime quality, and CFTC confidence when available. It is not a win probability.

Tradeability measures whether the market is clean enough to focus on. High tradeability should be validated through history; if it does not outperform, the model must be tuned.

Regime describes the market state: trend, range, coil, breakout, wash/rinse, volatile, or insufficient data. Mode decides whether the regime is actionable.

The history page stores frozen snapshots, COT z-scores, and later 1D / 5D / 10D market movement. This is where misses like GOLD/EURUSD can be measured objectively.

The report page summarizes what worked, what failed, regime mix, best and worst signals, and model-tuning notes.

Scores should be stored as they were at the time. The purpose is to improve Atlas honestly, not rewrite past signals.